2 x 30 Credits - Detection, modelling and tracking of pedestrians in heavy-duty autonomous vehicles i Sodertalje

Scania is now undergoing a transformation from being a supplier of trucks, buses and engines to a supplier of complete and sustainable transport solutions. Autonomous vehicle development at Scania is advancing at a very high pace and self-driving trucks and buses on public roads will soon see the light of day.

Thesis project at Scania is an excellent way of making contacts for your future working life. Many of our current employees started their career with a thesis project.

Background
Autonomous Transport Solutions (ATS) Research is responsible for developing, testing and piloting future frontier ATS concepts. This work is done using agile and self-steered teams with the ambition to detect and evaluate upcoming technologies, and prepare these for industrialization. We work in close cooperation with Volkswagen Group Innovation, leading technology suppliers and academic institutions.

One of the groups within the research department, EARP has the strategic and operational responsibility for the research of next generation Perception and Localization solutions for the ATS environment.

In this role, you will have the unique opportunity to be part of setting the future of Scania’s innovative Autonomous Transport Solutions. You will be part of a highly competent multicultural team instrumental in developing cutting-edge autonomous technologies where your ideas will be encouraged and embraced.

Project description
The development of autonomous vehicles relies in a clear understanding of the vehicle surroundings, as they are used to determine the driving actions to account for the existing risks in the environment. Therefore, accurate estimation methods are required in order to enable autonomous driving.

One of the main challenges for clearly understanding the vehicle surroundings is the detection, modelling and tracking of obstacles and road users in a dynamic environment. One of the most important road users to be considered are pedestrians, who play a crucial role for route planning and risk awareness in an autonomous vehicle.

To properly account for the movement of pedestrians with onboard sensors, an algorithm implemented for such purposes must: (i) properly detect the presence of pedestrians in the environment, (ii) model the possible movements of different pedestrian, and (iii) track the detected pedestrians in real time. The present project is focused on addressing the previous items.

This thesis project will explore the use of 3D lidar point clouds for pedestrian detection and tracking. To this end, the present project proposes the analysis and implementation of a detector algorithm making use of this sensor modality, as well as the analysis and implementation of mathematical models for pedestrian tracking.

Scope of the project
The objectives of this Master thesis project are

• Define and evaluate methods for pedestrian detection using 3D lidar point clouds.
• Define and evaluate dynamical models for pedestrian tracking using 3D lidar point clouds.

This project is divided into two master thesis proposals:

1. Detection of pedestrians using 3D lidar point clouds (1 master thesis student).
2. Modelling and tracking of pedestrians using 3D lidar point clouds (1 master thesis student).

Duration: 20 weeks
Start: January 2020
Credits: 30 HP X 2 (ECTS)

Contact persons and supervisors
Patricio E. Valenzuela, +46 8 553 722 45,
Navid Mahabadi, +46 8 553 727 26, 
Mansoureh Jesmani, +46 8 553 503 59,